Masked Face Detection and Recognition from Images

Abstract

The research field of computer vision has recently taken interest in the active problem of masked face recognition due to the COVID 19 pandemic. The use of face masks as a preventative measure against the spread of COVID-19 has presented a new challenge for the technology of face recognition. Masked Face Recognition (MFR) has emerged as a crucial issue within the field of face recognition after the COVID-19 epidemic. MFR is a specific type of facial occlusion issue that obstructs vital facial features such as the mouth, nose, or chin. The purpose of research on Masked Face Detection and Recognition is to fine-tune a pre-trained model that can more accurately recognize masked faces and detect whether the person is wearing a mask or not, which can be advantageous in various applications such as security and surveillance, healthcare, retail, law enforcement, the workplace, and social media. The objective of this dissertation is to examine the potential of machine learning techniques for enhancing the performance of masked face detection and recognition systems. This thesis proposes an approach to enhance the performance of the single neural network architecture such as pretrained InceptionV3 as unified model capable of both detection and recognition of masked images by achieving 99% and 98% respectively on MFR2 dataset. Pretrained VGG16 with transfer learning and fine tuning is trained and tested on publicly available datasets for the detection of masked faces, which are MDMFR dataset, Kaggle face mask detection dataset, facedatahybrid and for recognition results obtained on MFR2. The findings of this research offer valuable insights into the potential of pretrained networks with transfer learning to improve the performance of masked face detection and recognition systems and pave the way for future research in this area.